The proteasome, the most complex proteolytic assembly known, is a key mediator of biological regulation in eukaryotes. Ubiquitinated substrates are recognized by the 900-Kda regulatory particle (RP), then unfolded and translocated through a gated channel into the interior chamber of the core particle (CP) to be degraded. The key steps of recognition and unfolding are poorly understood, but appear to be critically dependent on a subassembly of the RP known as the base. The base is in direct contact with the CP, and is composed of 8 subunits, 6 of them ATPases of the AAA family. The present proposal aims to better understand how the base promotes protein degradation. Substrate recognition appears to be mediated by an ensemble of ubiquitin chain binding factors, including Rpnl0 and the UU proteins (such as Rad23 and Dsk2), all of which associate with the base. Still other factors, possibly intrinsic to the base, are also likely to be involved. In Aim 1 we will take a combined genetic, biochemical, and proteomic approach to resolve how these multiple pathways of substrate recognition are related: do they function redundantly, cooperatively, or independently? Are their functions overlapping for some substrates while being uniquely required to recognize others? The main focus of Aim 2 is an almost completely unstudied part of the base, the N-terminal domains of the 6 ATPases. Our hypothesis is that the N-domains play key roles in protein degradation, by projecting into the central compartment of the RP, where unfolding is thought to occur, and helping to promote this event by interacting directly with substrate. Among our proposed tests of this idea are attempts to screen genetically for mutations in the N-domains that differentially impair the degradation of one substrate but not another, at a step (presumably unfolding) that follows ubiquitin chain recognition. Aim 3 addresses our hypothesis that the largest subunit of the base, Rpnl, acts as a scaffold to recruit multiple factors to the proteasome, all of which are active on multiubiquitin chains. The proposed multiple functions of Rpnl will be dissected mainly by obtaining point mutants that specifically abrogate individual binding sites for postulated ligands such as Rad23, Rpnl0, and Ubp6. In summary, this proposal will combine biochemical, genetic, proteomic, and structural methods to address key questions concerning early steps in protein degradation by the proteasome.